UCT/IB/MD: retry memory registration with reduced access flags on failure #10341
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -4,6 +4,7 @@ Alex Margolin <[email protected]> | |
| Alex Mikheev <[email protected]> | ||
| Alexey Rivkin <[email protected]> | ||
| Alina Sklarevich <[email protected]> | ||
| Alma Mastbaum <[email protected]> | ||
| Anatoly Vildemanov <[email protected]> | ||
| Andrey Maslennikov <[email protected]> | ||
| Artem Polyakov <[email protected]> | ||
|
|
||
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -309,7 +309,7 @@ void *uct_ib_md_mem_handle_thread_func(void *arg) | |
| while (ctx->length > 0) { | ||
| if (ctx->params != NULL) { | ||
| status = uct_ib_reg_mr(ctx->md, ctx->address, length, ctx->params, | ||
| ctx->access_flags, NULL, &ctx->mrs[mr_idx], 1); | ||
|
There was a problem hiding this comment. why hide_errors is true here? There was a problem hiding this comment. @yosefe Oh, I actually wanted to put here a value that doesn't change the previous behavior and put |
||
| if (status != UCS_OK) { | ||
| goto err_dereg; | ||
| } | ||
|
|
@@ -460,7 +460,7 @@ uct_ib_md_handle_mr_list_mt(uct_ib_md_t *md, void *address, size_t length, | |
| ucs_status_t uct_ib_reg_mr(uct_ib_md_t *md, void *address, size_t length, | ||
| const uct_md_mem_reg_params_t *params, | ||
| uint64_t access_flags, struct ibv_dm *dm, | ||
| struct ibv_mr **mr_p, int first_attempt) | ||
|
There was a problem hiding this comment.
|
||
| { | ||
| ucs_time_t UCS_V_UNUSED start_time = ucs_get_time(); | ||
| unsigned long retry = 0; | ||
|
|
@@ -507,7 +507,8 @@ ucs_status_t uct_ib_reg_mr(uct_ib_md_t *md, void *address, size_t length, | |
| if (mr == NULL) { | ||
| uct_ib_md_print_mem_reg_err_msg(title, address, length, access_flags, | ||
| errno, | ||
| (flags & UCT_MD_MEM_FLAG_HIDE_ERRORS) || | ||
| first_attempt); | ||
| return UCS_ERR_IO_ERROR; | ||
| } | ||
|
|
||
|
|
@@ -626,9 +627,39 @@ ucs_status_t uct_ib_memh_alloc(uct_ib_md_t *md, size_t length, | |
| return UCS_OK; | ||
| } | ||
|
|
||
| uint64_t uct_flags_to_ibv_mem_access_flags(uint64_t uct_flags) | ||
|
|
||
| { | ||
| uint64_t ibv_flags = 0; | ||
|
|
||
| if (uct_flags & UCT_MD_MEM_ACCESS_LOCAL_WRITE) { | ||
| ibv_flags |= IBV_ACCESS_LOCAL_WRITE; | ||
| } | ||
|
|
||
| if (uct_flags & UCT_MD_MEM_ACCESS_REMOTE_GET) { | ||
| ibv_flags |= IBV_ACCESS_REMOTE_READ; | ||
| } | ||
|
|
||
| if (uct_flags & UCT_MD_MEM_ACCESS_REMOTE_PUT) { | ||
| ibv_flags |= IBV_ACCESS_REMOTE_WRITE; | ||
| } | ||
|
|
||
| if (uct_flags & UCT_MD_MEM_ACCESS_REMOTE_ATOMIC) { | ||
| ibv_flags |= IBV_ACCESS_REMOTE_ATOMIC; | ||
| } | ||
|
|
||
| return ibv_flags; | ||
| } | ||
|
|
||
| uint64_t uct_ib_memh_access_flags(uct_ib_mem_t *memh, int relaxed_order, | ||
| int first_attempt, uint64_t uct_flags) | ||
|
There was a problem hiding this comment. Here as well, I'd rename |
||
| { | ||
| uint64_t access_flags; | ||
|
|
||
| if (first_attempt) { | ||
| access_flags = UCT_IB_MEM_ACCESS_FLAGS; | ||
| } else { | ||
| access_flags = uct_flags_to_ibv_mem_access_flags(uct_flags); | ||
| } | ||
|
|
||
| if (memh->flags & UCT_IB_MEM_FLAG_ODP) { | ||
| access_flags |= IBV_ACCESS_ON_DEMAND; | ||
|
|
@@ -645,26 +676,33 @@ ucs_status_t uct_ib_verbs_mem_reg(uct_md_h uct_md, void *address, size_t length, | |
| const uct_md_mem_reg_params_t *params, | ||
| uct_mem_h *memh_p) | ||
| { | ||
| uct_ib_md_t *md = ucs_derived_of(uct_md, uct_ib_md_t); | ||
| uint64_t uct_flags = UCT_MD_MEM_REG_FIELD_VALUE(params, flags, | ||
| FIELD_FLAGS, 0); | ||
| int attempt_cnt = 0; | ||
| struct ibv_mr *mr_default; | ||
| uct_ib_verbs_mem_t *memh; | ||
| uct_ib_mem_t *ib_memh; | ||
| uint64_t access_flags; | ||
| ucs_status_t status; | ||
|
|
||
| status = uct_ib_memh_alloc(md, length, uct_flags, | ||
| sizeof(*memh), sizeof(memh->mrs[0]), &ib_memh); | ||
| if (status != UCS_OK) { | ||
| goto err; | ||
| } | ||
|
|
||
| memh = ucs_derived_of(ib_memh, uct_ib_verbs_mem_t); | ||
|
|
||
| do { | ||
| access_flags = uct_ib_memh_access_flags(&memh->super, md->relaxed_order, | ||
| attempt_cnt == 0, uct_flags); | ||
|
|
||
| status = uct_ib_reg_mr(md, address, length, params, access_flags, NULL, | ||
|
|
||
| &mr_default, attempt_cnt == 0); | ||
| attempt_cnt++; | ||
| } while ((status != UCS_OK) && (attempt_cnt < 2)); | ||
|
|
||
| if (status != UCS_OK) { | ||
| goto err_free; | ||
| } | ||
|
|
@@ -676,7 +714,7 @@ ucs_status_t uct_ib_verbs_mem_reg(uct_md_h uct_md, void *address, size_t length, | |
| if (md->relaxed_order) { | ||
| status = uct_ib_reg_mr(md, address, length, params, | ||
| access_flags & ~IBV_ACCESS_RELAXED_ORDERING, | ||
| NULL, &memh->mrs[UCT_IB_MR_STRICT_ORDER].ib, 1); | ||
| if (status != UCS_OK) { | ||
| goto err_dereg_default; | ||
| } | ||
|
|
||
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Needed?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It's just the most basic option so I thought I can keep mention it as well, but on the other hand it doesn't matter as local read is enabled by default in ibv_reg_mr (there is no such flag in verbs.h because it's always enabled)